05/29/2018 Tue Lab Meeting

Post date: May 29, 2018 5:11:27 PM

Attendees: 오동훈, 이장원, 박진희, 장주은

• Donghoon find a way to read HUGE csv file

  • Chop the large one into 10 MB chunks and read in data in small unit

  • Use python pandas read CSV features

• Field values have different types

  • Strings

  • Numbers

  • Ranges (e.g. ">0.01")

• Find unique values, and apply a good embedding method